%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Data for experiments in the paper: Jet Substructure Classification in High-Energy Physics with Deep Neural Networks Pierre Baldi, Kevin Bauer, Clara Eng, Peter Sadowski, Daniel Whiteson http://arxiv.org/abs/1603.09349 Date: April 9th 2016 Author: Peter Sadowski peter.j.sadowski@uci.edu %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Data set consists of 10M training examples, 5M test examples. There are two sets of features: the low-level calorimeter images, and the high-level derived features. There are also two versions of the datasets, one with pile-up and one without pile-up.